fastlowess-wasm

One LOWESS to Rule Them All

The fastest, most robust, and most feature-complete language-agnostic LOWESS (Locally Weighted Scatterplot Smoothing) implementation for Rust, Python, R, Julia, Node.js, C++, Go, Java, and WebAssembly.

The lowess-project also offers bindings for Rust, Python, R, Julia, Node.js, WebAssembly, C++, Go, and Java — see the full repository.

---

Installation & Documentation

Currently available for R, Python, Rust, Julia, Node.js, WebAssembly, and C++. See the Installation Guide for detailed installation instructions.

📚 View the full documentation

---

LOESS vs. LOWESS

Feature	LOESS	LOWESS (This Crate)
Polynomial Degree	Linear, Quadratic, Cubic, Quartic	Linear (Degree 1)
Dimensions	Multivariate (n-D support)	Univariate (1-D only)
Flexibility	High (Distance metrics)	Standard
Complexity	Higher (Matrix inversion)	Lower (Weighted average/slope)

Read more about how LOWESS works in the Concepts.

Note: For a LOESS implementation, use loess-project.

---

Why this package?

Speed

The shared native benchmarks report that the project is on average 200-327x faster than Python’s statsmodels.lowess and up to 7.8× faster than base R’s stats::lowess on a tested large, wide-window workload. These results do not measure the single-threaded WebAssembly binding; see the Benchmarks page for details.

Robustness

This implementation is more robust than R’s lowess and Python’s statsmodels due to two key design choices:

MAD-Based Scale Estimation:

For robustness weight calculations, this crate uses Median Absolute Deviation (MAD) for scale estimation:

s = median(|r_i - median(r)|)

In contrast, statsmodels and R’s lowess uses the median of absolute residuals (MAR):

s = median(|r_i|)

• MAD is a breakdown-point-optimal estimator—it remains valid even when up to 50% of data are outliers.
• The median-centering step removes asymmetric bias from residual distributions.
• MAD provides consistent outlier detection regardless of whether residuals are centered around zero.

Boundary Padding:

This crate applies a range of different boundary policies at dataset edges:

• Extend: Repeats edge values to maintain local neighborhood size.
• Reflect: Mirrors data symmetrically around boundaries.
• Zero: Pads with zeros (useful for signal processing).
• NoBoundary: Original Cleveland behavior

statsmodels and R’s lowess do not apply boundary padding, which can lead to:

• Biased estimates near boundaries due to asymmetric local neighborhoods.
• Increased variance at the edges of the smoothed curve.

Features

A variety of features, supporting a range of use cases:

Feature	This package	statsmodels	R (stats)
Kernel	7 options	only Tricube	only Tricube
Robustness Weighting	3 options	only Huber	only Huber
Scale Estimation	2 options	only MAR	only MAR
Boundary Padding	4 options	no padding	no padding
Zero Weight Fallback	3 options	no	no
Auto Convergence	yes	no	no
Online Mode	yes	no	no
Streaming Mode	yes	no	no
Confidence Intervals	yes	no	no
Prediction Intervals	yes	no	no
Cross-Validation	2 options	no	no
Parallel Execution	yes	no	no
no-std Support	yes	no	no

Validation

All implementations are numerical twins of R’s lowess:

Aspect	Status	Details
Accuracy	✅ EXACT MATCH	Max diff < 1e-12 across all scenarios
Consistency	✅ PERFECT	Multiple scenarios pass with strict tolerance
Robustness	✅ VERIFIED	Robust smoothing matches R exactly

---

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for more information.

License

Licensed under MIT or Apache-2.0.

Citation

If you use this software in your research, please cite it using the CITATION.cff file or the BibTeX entry below:

@software{lowess_project,
  author = {Valizadeh, Amir},
  title = {LOWESS Project: High-Performance Locally Weighted Scatterplot Smoothing},
  year = {2026},
  url = {https://github.com/thisisamirv/lowess-project},
  license = {MIT OR Apache-2.0}
}